Modified prophylactic 5-fr pancreatic duct stent enhances the rate of spontaneous dislodgement: A multicenter randomized controlled trial

Abstract

Background and objectives

Prophylactic pancreatic duct stent placement effectively reduces post-endoscopic retrograde cholangiopancreatography pancreatitis in high-risk patients, but the optimal stent remains unclear. We modified a 5-Fr, 3 cm pancreatic stent by removing the flange on the pancreatic side and compared the rate of spontaneous dislodgement and complications with the ordinary stent.

Methods

This was a randomized controlled trial at six tertiary endoscopic centers. Patients deemed high risk for post-endoscopic retrograde cholangiopancreatography pancreatitis randomly received modified or ordinary pancreatic stent. The primary outcome was spontaneous stent dislodgement at five days and 14 days. Secondary outcomes were the success rate of stent placement and complications.

Results

A total of 276 patients were randomly assigned to receive modified stents (mS group) and ordinary stents (oS group). The placement of a pancreatic stent was successful in all 276 patients. There were no significant differences between groups with respect to age, sex, major diagnosis, or indications for stenting. At five days the spontaneous dislodgement rate was 47.72% for the mS group and 15.67% for the oS group (p<0.001); at 14 days the rates were 84.21% and 42.65%, respectively (p < 0.001). Post-endoscopic retrograde cholangiopancreatography pancreatitis occurred in 6.52% of all patients. There were no significant differences regarding the incidences of post-endoscopic retrograde cholangiopancreatography pancreatitis, hemorrhage or fever.

Conclusions

The modified short 5-Fr stent has a higher spontaneous dislodgement rate than ordinary pancreatic stent, thus obviating the need for endoscopic removal. The modified pancreatic stent does not increase the incidence of post-endoscopic retrograde cholangiopancreatography pancreatitis or other complications. The endoscopist can consider removing the flange on the pancreatic duct side for prophylactic pancreatic duct manipulation.

Key summary

Temporary prophylactic pancreatic stent (PS) placement has been shown to reduce post-endoscopic retrograde cholangiopancreatography pancreatitis (PEP) in high-risk patients. A short single-pigtail 5-Fr PS which has a flange on the pancreatic ductal (proximal) side is widely used in China. The spontaneous dislodgement for this kind of short 5-Fr PS with proximal flange is relatively low.

We modified the PS by removing the flange on the proximal side. The modified PS enhanced the rates of spontaneous passage from 42.65% to 84.21% at 14 days. The modified PS was easily placed and did not increase the level of adverse events.

Introduction

Pancreatitis is the most common complication of endoscopic retrograde cholangiopancreatography (ERCP) and is commonly referred to as post-ERCP pancreatitis (PEP).¹ A systematic survey of 21 prospective studies with 16,855 patients conducted between 1987–2003 found a 3.5% occurrence rate of PEP, among which 0.4% of patients developed severe pancreatitis, resulting in a 0.11% death rate.² Given the high magnitude of this problem, the prophylaxis of PEP is crucial to the outcome of ERCP.

Our previous retrospective single-center analysis of 4234 ERCP procedures between September 2007–December 2012 found that PEP occurred in 226 (5.3%) patients and risk factors included female sex, first time ERCP, latent jaundice, difficult cannulation, and pancreatography. To prevent PEP in these high-risk patients, prophylactic pancreatic duct (PD) stenting proved to be useful.³ Other studies have also reported that temporary prophylactic pancreatic stent (PS) placement is effective in reducing PEP in high-risk patients^4–6 and is cost-effective.⁷ Consistent with our results, a more recent meta-analysis of 1541 patients found that pancreatic stenting prevented PEP after ERCP compared with no PS placement, also supporting the importance of PS placement after ERCP for the prevention of PEP.⁸

The prophylactic PS commonly includes straight PS and single-pigtail PS with a caliber of 3 Fr or 5 Fr. Compared to the straight PS, the single-pigtail PS minimizes stent proximal migration and is safe. The placement of a 3-Fr stent is more challenging as it requires a smaller caliber 0.018-inch wire, which is harder to manoeuver and visualize under fluoroscopy,⁹ and is not available in many countries. However, the prophylactic 5-Fr PS is considered to be the effective stent for preventing PEP, with fewer complications and a relatively low rate of spontaneous dislodgement. One should consider that PD stent retention for prolonged periods of time is a significant risk factor for chronic pancreatitis.¹⁰

A short 5-Fr single-pigtail stent that is characterized by a radiopaque marker and one flange on the pancreatic ductal (proximal) side is commonly used in China, but the rate of spontaneous dislodgement remains at approximately 65%.⁸ To increase the rate of spontaneous dislodgement, we modified the PS by removing the flange on the proximal side. The primary aim of this multicenter randomized study was to compare the spontaneous dislodgement of modified and ordinary PSs. Our secondary aim was to assess and compare the success of stent placement and complications between two groups.

Material and methods

Study design

All consecutive patients undergoing ERCP with the intent to cannulate the common bile duct (CBD) were considered eligible for recruitment. Written informed consent was obtained from every patient prior to ERCP. Inclusion criteria were (a) intent to cannulate the CBD; (b) adults ( ≥ 18 years); and (c) difficult cannulation and PD cannulation was conducted. A difficult bile duct cannulation was defined as when any of the following conditions were met: (a) the time of cannulation was more than five minutes; (b) accidental PD passage more than three times; and (c) a PD fluoroscopy with contrast. Exclusion criteria were (a) ERCP previously performed with sphincterotomy, endoscopic papillary dilation or stent insertion; (b) a mass of papilla; (c) previous surgeries, including gastrointestinal bypass or choledochoenterostomy; (d) upper gastrointestinal stricture; (e) acute pancreatitis before ERCP; (f) coagulation disorders, heart failure or renal failure; and (g) suspected pancreatic head carcinoma. Patients were enrolled from six public tertiary referral endoscopic centers between September 2015–July 2017.

This was a multicenter, prospective, randomized, controlled clinical trial. Randomization was conducted by the study designer using per the block randomization method.¹¹ Patients were assigned to the following treatment groups: the modified stent group (mS group) and ordinary stent group (oS group). Allocation was enclosed in an opaque sealed envelope, which was revealed when the endoscopist decided to place a prophylactic PS.

The data collected included the patients’ baseline characteristics, indications for ERCP, indications of difficult cannulations, post-procedural PS dislodgements and adverse events. The primary endpoint was the spontaneous dislodgement rate of the modified and ordinary PSs at five days and 14 days. The secondary endpoints included the success rate of stent placement and complications.

ERCP procedure

Ten experienced endoscopists who had performed more than 1000 ERCPs conducted all of the procedures. The intravenous administration of Propofol with remifentanil was used to achieve unconscious sedation. ERCP was performed in a standard fashion with JF-260 V or TJF-240 endoscopes (Olympus, Tokyo, Japan). Cannulation was conducted using a disposable papillotomy knife (CleverCut KD-211Q-0725, OLYMPUS, Tokyo, Japan) with a preloaded hydrophilic guidewire (Radifocus guide wire m, RF-GA35263M TERUMO, Vietnam) and then exchanged with Jagwire (Straight tip, M00556580, Boston Scientific, Costa Rica) after deep cannulation was successful. In cases of difficult cannulation, when a hydrophilic guidewire could be inserted into the PD, a Jagwire was exchanged, using double guidewire technology to assist the bile duct cannulation. After accomplishing bile duct cannulation, the following endoscopic procedures were performed: sphincterotomy, bile duct stone removal, biliary stent placement or nasobiliary drainage. The timepoint of PS placement was determined by the endoscopist. The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in prior approval by the institution's human research committee.

Pancreatic stents

The PD stent used in our study was a 5-Fr 3 cm stent, with a single pigtail on the duodenal side (Zimmon, SPSOF-5-3, Cook Medical, Limerick, Ireland). This stent is characterized by a radiopaque marker with one flange on the pancreatic ductal side. For the modified stent group, the flange was removed using a surgical blade (Figure 1).

Figure 1.

Removal of flange in modified stent.

Follow-up

After the ERCP procedure, patients were admitted to the ward for five days. The symptoms, physical signs, amylase levels and full blood analyses were recorded. After discharge, patients were followed up in the outpatient clinic or by telephone until the trial was finished. All patients were informed of the need to observe whether the PD was defecated with their bowel movements. The patients without reports of PD stent dislodgement were evaluated by abdominal X-ray to check for stent migration at five days and 14 days after the procedure. In cases in which the stent had not passed by 14 days, the stents were removed endoscopically.

Definitions

PEP was defined by consensus on the following criteria: the occurrence of intense and continuous abdominal pain, an elevation in pancreatic amylase to three times the upper limit of normal at 24 h after the procedure, and hospital admission longer than 10 days (severe pancreatitis).

Post-ERCP bleeding was defined as the presence of melena with or without hemodynamic compromise, a decrease in the value of hemoglobin by at least 2 g/dl, or the need for a red blood cell transfusion after the procedure. The requirement for a blood transfusion or angiographic therapy was classified as moderate in severity. Bleeding was rated as severe if it resulted in a prolonged hospital stay ( > 10 days), resulted in a prolonged intensive care unit stay (>1 day) or required surgical intervention.

Perforation was defined as the presence of pneumoperitoneum or retropneumoperitoneum by an adequate imaging test with the accompanying clinical presentation.

Statistics

This sample size was calculated to have 80% statistical power to detect a non-inferiority margin difference between the two groups. The reference group proportion is 65%. The treatment group proportion is assumed to be 65% under the null hypothesis of inferiority. The power was computed for the case when the actual treatment group proportion is 80%. The test statistic used was the one-sided Z test (unpooled). The significance level of the test was targeted at 0.0250. The two-tailed Student’s t-test was used to assess for differences between the continuous means of the two groups. The two-tailed Fisher’s exact and chi-square tests were used to compare categorical variables. A value of p < 0.05 was considered to be statistically significant. All statistical analyses were performed using SPSS software package, version 21 (Statistical Package for Social Sciences, IBM Corporation, Armonk, New York, USA).

Ethics considerations

The Ethics Committee of Nanjing Drum Tower Hospital of Nanjing University Medical School, Nanjing, China, approved the study on 23 June 2015.

Results

Patient and procedure data

During the study period, a total of 2874 patients passed the initial screening in the six endoscopic centers, of whom 1494 were excluded according to the exclusion criteria. After initial exclusion, 998 patients had not undergone difficult cannulations, as described above, and 106 patients declined to consent for participation in the study, leading to a total of 276 patients enrolled in the study. These patients were randomly assigned to receive modified stents (n = 138) or ordinary stents (n = 138). Every center was required to complete 46 cases. See Figure 2 for details on screening and enrolment. The demographic parameters and indications for ERCP did not significantly differ between the two groups (Table 1). The risks of PEP and elevated bilirubin revealed no statistically significant differences (Table 1). Table 2 shows the technical characteristics of the modified and ordinary stent groups. There were no differences between groups in terms of the number of patients who underwent difficult cannulations, endoscopic nasobiliary drainage (ENBD), biliary stent placement, or other technical characteristics. All patients were admitted for five days after the procedure, and 273 patients completed the five-day and 14-day post-ERCP follow-ups.

Figure 2.

Details of screening and enrolment. ERCP: endoscopic retrograde cholangiopancreatography; AP: acute pancreatitis; mS group: modified stent group; oS group: ordinary stent group.

Table 1.

Demographics and indications for endoscopic retrograde cholangiopancreatography (ERCP).

	mS Group (%)	oS Group (%)	p
No. of patients	138 (100)	138 (100)
Sex
Male	74 (53.6)	68 (49.3)
Female	64 (46.4)	70 (50.7)	0.470
Age (years)	58 ± 15	60 ± 16	0.290
Major diagnosis
CBD stones	104 (75.4)	105 (76.1)	0.678
Bile duct benign stenosis	6 (4.3)	3 (2.2)
Malignant obstruction of CBD	14 (10.1)	16 (11.6)
Hilar cholangiocarcinoma	7 (5.1)	3 (2.2)
Bile leakage	1 (0.7)	2 (1.4)
Others	6 (4.3)	8 (5.8)
Elevated bilirubin	103 (74.6)	92 (66.7)	0.146
History of acute pancreatitis	19 (13.8)	29 (21)	0.112

CBD: common bile duct; mS group: modified stent group; oS group: ordinary stent group.

Table 2.

Summary of technical characteristics.

	mS Group (%)	oS Group (%)	p
Difficult cannulation
Cannulation time >5 min	76 (55.9)	68 (49.6)	0.335
Cannulation >3 attempts	93 (68.4)	97 (70.8)	0.603
PD guidewire passage  >3 attempts	68 (50)	68 (49.6)	1
Pancreatography	15 (11)	14 (10.2)	0.844
ENBD	101 (74.3)	108 (78.8)	0.326
BD stent placement	15 (11)	18 (13.1)	0.578
Sphincterotomy	110 (80.9)	118 (86.1)	0.204
BD stone removal	105 (77.2)	111 (81)	0.381
Rectal indomethacin	6 (4.4)	12 (8.8)	0.144

ENBD: endoscopic nasobiliary drainage; BD: bile duct; mS group: modified stent group; oS group: ordinary stent group; PD: pancreatic duct.

Outcomes of pancreatic stent

Table 3 shows the outcomes of the PD. The prophylactic PS was successfully placed in all 276 patients; three stents in the mS group and one stent in oS group spontaneously fell out during the ERCP procedure and were not replaced (none of these four patients presented with PEP). A total of three stents (one in the mS group and two in the oS group) were removed endoscopically by five days; two were removed to stop bleeding, and one was removed to exchange the bile duct stent. A significant difference in the rate of stent spontaneous dislodgement at five days was detected between the mS group and oS group (mS group, n = 63; oS group, n = 21; p < 0.001). From the sixth day to the thirteenth day, two stents in each group were removed with an endoscopic procedure or surgery. The 14-day spontaneous dislodgement rates in the mS and oS groups were 112 (84.21%) and 58 (42.65%), respectively (p < 0.001). The patients (mS group, n = 18; oS group, n = 64;) without stent spontaneous dislodgement at 14 days underwent endoscopic removal procedures.

Table 3.

Outcomes of pancreatic stent.

	mS Group (%)	oS Group (%)	p
Successful placement	138 (100%)	138 (100%)	1
Dislodged during ERCP	3 (2.17%)	1 (0.07%)	0.614
Lost to follow-up	2 (1.48%)	1 (0.07%)	1
Removed in five days	1 (0.75%)	2 (1.47%)	1
Spontaneous dislodgement within 5 daysa	63 (47.72%)	21 (15.67%)	<0.001
Removed within 6∼13 daysb	2 (2.90%)	2 (1.77%)	1
Spontaneous dislodgement within 6∼14 days	49 (73.13%)	37 (33.33%)	0.119
Spontaneous dislodgement within 14 days	112 (84.21%)	58 (42.65%)	<0.001

ERCP: endoscopic retrograde cholangiopancreatography mS group: modified stent group; oS group: ordinary stent group.

^aIn the mS group, 21 patients reported that the stents were defecated, and the other 42 stent dislodgements were confirmed by abdominal X-ray. In the oS group, the numbers of stents reported and confirmed X-rays were three and 18, respectively.

^bBetween the sixth and thirteenth days, two stents in the mS group were removed during an endoscopic procedure, while two stents in the oS group were removed (one during surgery and one by endoscopy).

Adverse events

Table 4 shows the incidence rates of all complications, including PEP, fever, hemorrhage and death, in the two groups. A total of 18 patients (6.52%) met the criteria of PEP. Only mild or moderate PEP was observed. Of the seven patients in the mS group (5.07%) who developed PEP, one had moderate PEP. Of the 11 patients in the oS group (7.97%) who developed PEP, two had moderate PEP. There was no significant difference between the PEP rates of the mS and oS stent groups (p = 0.329). Ten patients presented with fever after the ERCP procedure, occurring in seven patients in the mS group and three patients in the oS group (p = 0.198). Among the 10 patients who had fever, six (three in the mS group and three in the oS group) were treated with stent exchange or surgery, and the others only received antibiotics. Three patients in the mS group and six patients in the oS group developed post-sphincterotomy bleeding; three patients had moderate bleeding with one in the mS group requiring a blood transfusion and two in the oS group requiring endoscopic hemostasis. There were no significant differences between the two groups regarding the complications described above.

Table 4.

Overall complications.

	mS Group (%)	oS Group (%)	p
Post-ERCP pancreatitis	7 (5.07%)	11 (7.97%)	0.329
Fever	7 (5.07%)	3 (2.17%)	0.198
Hemorrhage	3 (2.17%)	6 (4.35%)	0.501
Death	1 (0.72%)	1 (0.72%)	1

ERCP: endoscopic retrograde cholangiopancreatography; mS group: modified stent group; oS group: ordinary stent group.

Discussion

The various mechanisms of PEP include the post-procedural impaired drainage of the PD caused by papillary enema, spasm of the sphincter of Oddi, or PD injury as a result of the procedure, such as repetitive contrast injections.^5,10 The positioning of a temporary PS across the ampulla and pancreatic sphincter into the PD is presumed to maintain the flow of pancreatic secretions across any flow disruption due to the injury or edema of these structures. A meta-analysis has shown that PS placement decreases the risk of PEP and hyperamylasemia in high-risk patients.¹² The results of our study are generally consistent with those of other reports.^{4–6,8,13–18}

Stents with various diameters (3 F–7 F), with various lengths (3.0–12 cm), and with or without flanges have been used in previous studies of PEP. Zolotarevsky et al.¹⁹ reported that the placement of 5-Fr PSs for PEP prophylaxis is easier, faster, and requires fewer wires than the placement of 3-Fr stents. There were no statistically significant differences in the spontaneous passage rates (5-Fr group, 68.4%; 3-Fr group, 75.0%; p = 0.617). Furthermore, among patients at high-risk for PEP, the spontaneous dislodgement rate of unflanged, short-length, 5-Fr PD stents is significantly higher than that of unflanged, long-length, 3-Fr stents (5-Fr stents 98% and 3-Fr stents 88%)⁹. In contrast, Rashdan et al.²⁰ reported that 3 F–4 F stents are more effective than traditionally used stents for the prevention of PEP and that 5 F–6 F stents cause more significant stent-induced PD changes than 3 F–4 F stents. However, the placement of a 3 F–4 F stent requires a small-caliber guidewire (0.018–0.025 in), and the procedure for handling a small-caliber guidewire is difficult and requires a high level of experience. Furthermore, the 3 F–4 F stent and small-caliber guidewire are not available in China. Based on our experience in a preliminary study,³ we selected a stent that was 5 Fr in diameter, 3 cm in length, had one pigtail and had a flange on the internal surface to use in the prophylaxis of PEP. We also considered that the one-pigtail type of PS shape on the duodenal side was easier to dislodge as a result of friction with passing foods and duodenal peristalsis and more difficult to migrate proximally into the PD.

Although a short 5-Fr PS is effective in preventing PEP, the rate of spontaneous dislodgement remains low. Whether the rate of spontaneous passage will be increased if the internal flange is removed is unclear. The present study showed that the absence of internal flanges is important for spontaneous passage. We measured the spontaneous passage rate by calculating the proportion of patients who had spontaneous passages of the PD stent at five days and 14 days after placement. The rates of spontaneous passage of modified 5-Fr stents were significantly higher than those of the ordinary stents (47.72% vs 15.67% at five days and 84.21% vs 42.65% at 14 days). Our results regarding the spontaneous passage of the 5-Fr stent were within the range of the rates reported in the above two studies (67% and 98%, respectively).^9,20

Experientially, some technical factors during ERCP, which include sphincterotomy, bile duct stent placement, and ENBD, may influence the spontaneous passage of the PS. In our study, there were no statistically significant differences in these characteristics between the mS group and oS group. These findings suggest that the flange on the PD side influences the spontaneous passage of the PS. An important technical factor, which may influence the type of temporary prophylactic PS chosen, is the technical ease of placement because prolonged and unsuccessful attempts at placement may potentially increase the risk of developing PEP.²¹ In our series, the 5-Fr PS was successfully placed for the prophylaxis of PEP in all cases. Four PD stents dislodged during ERCP without subsequent stent replacement due to concerns regarding the increased risk of PEP associated with injury to the duodenal papilla and PD. None of these cases developed to PEP.

PS retention for a prolonged period of time is a significant risk factor with respect to chronic pancreatitis.¹⁰ Therefore, as in studies by other researchers, we removed the PS endoscopically if it did not dislodge within 14 days.^9,20

As evident from the indications for stent placement, our patients were considered to be at high risk for the development of PEP. In our study, the rates of PEP were comparable to those of the previous literature in high-risk patient populations with PD stent placement.^5,22 This study comes from tertiary endoscopic centers with extensive ERCP experience. Further studies in less-experienced centers are needed to determine more widespread applicability of modified prophylactic pancreatic stenting.

In conclusion, we found a significantly higher spontaneous dislodgement rate for the short 5-Fr stent without the flange than for the unmodified stent, thereby obviating the need for endoscopic removal at 14 days. Thus, an endoscopist may consider removing the flange on the PD side for prophylactic pancreatic duct manipulation.

Declaration of conflicting interests

None declared.

Ethics approval

The study was reviewed and approved by the ethics committee of the Affiliated Drum Tower Hospital of Nanjing University Medical School.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Informed consent

Informed consent was obtained from each participant.